The Agent Skills Directory

[SAFE]: The skill package is composed entirely of markdown documentation and reference files (SKILL.md, references/experiments.md, and references/form.md). There are no executable scripts, system commands, or remote code patterns identified.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructs the agent to check for project-specific marketing context in files like .agents/product-marketing.md. This is a standard practice for maintaining project context and does not involve unauthorized access to system files, environment variables, or sensitive credentials.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze marketing pages and forms which are externally provided, creating a potential ingestion point for untrusted data. 1. Ingestion points: External marketing page content and URLs provided by the user. 2. Boundary markers: None explicitly defined to separate untrusted content. 3. Capability inventory: The skill is limited to diagnostic analysis and recommendation generation; it lacks any tools for network operations, file writing, or command execution that could be exploited by injected instructions. 4. Sanitization: No sanitization is performed on ingested content, but the lack of exploitable actions renders this surface safe.

cro